Summa

Md. Asif Uddin

Deep learning researcher — vision, medical imaging, vision-language models, causal inference, bioinformatics and gene sequence analysis

Correlation is cheap. Only an intervention costs anything.
— the author

Dhaka, Bangladesh

Tabula IIThe School of AthensRaphael, 1509–1511The School of Athens is a masterpiece of the Renaissance that celebrates human knowledge, reason, philosophy, and the pursuit of truth.Wikimedia Commons · Public domain

Prologue

I spend most of my time trying to understand AI and learn how to build useful AI systems. I work with different models, train and fine-tune them, modify existing architectures, and sometimes build new approaches when a problem needs it. I enjoy taking an idea, understanding how it works, and then trying to apply it to a real problem.

My undergraduate thesis, HierarchiRetina, is one example of this, and it is now under review at the IEEE Journal of Biomedical and Health Informatics, with the preprint on arXiv. It is a cascade that screens retinal photographs for diabetic retinopathy, segments the lesions and vessels, and grades the eye, evaluated end to end so that every case missed at the first step counts as an error. I have worked on the full process, from finding and preparing the data to training and evaluating the models. Getting the data was also a major part of the work. I visited hospitals, contacted researchers and institutions, and went through more than 37 published papers looking for suitable datasets. Most requests were declined, but the authors of FGADR responded and allowed us to use their dataset.

Alongside my thesis, I work with my advisors on different research problems. One project on open-world species recognition has already been completed and submitted for publication. Another project focuses on kidney disease. We are working with CT scan images to develop automated analysis and generate reports from the scans. These projects have helped me understand how AI can be adapted to very different problems and domains.

More recently, I have started working on a problem in computational biology. Instead of only asking which genes are related to Alzheimer’s disease, the work looks at what might happen to a microglial cell when a specific gene is knocked down. We then compare the model’s predictions with experimental CRISPR screens. I find this interesting because it brings together AI, biology, and experimental research.

Next is an MSc in Computer Science and Engineering at BRAC University, starting in Fall 2026. My proposal has already been accepted. I want to continue doing research while also building a career in AI and machine learning. I do not see research and industry as two separate paths. I want to keep learning through research while building AI systems that can solve real problems.

Outside computer science, I have broad interests in philosophy, law, public policy, science, and society. I also love cooking and trying new recipes. I enjoy learning about how ideas shape the way we understand the world, how technology affects people, and how we can use what we learn to build things that are genuinely useful.

Compendium

Plate I2026

HierarchiRetina

Diabetic retinopathy screening and grading, evaluated end to end

Under review

Plate II2026

Rank Radii Transfer as Quantiles

Placing unseen species inside the taxonomy, not in flat clusters

Under review

Plate III2026

Automated Renal Reporting

The aim: a generator that cannot emit a value it did not measure

In preparation

Plate IV2026

HCGT-PG

What would happen if you knocked the gene down

Proposal accepted

Tabula XXIISocratesLysippos, The ancient Greek philosopher Socrates was born around 470/469 BC and died in 399 BC at about 71 years old in Athens, Greece.Socrates spent his life asking questions that made other people uncomfortable. He did not build a school, write a book, or claim to possess wisdom. Instead, he walked through Athens asking people what justice, courage, virtue, and knowledge really meant. In the end, Athens executed him for doing exactly that. His life left behind one of philosophy's most disturbing questions: what if a life without self-examination is not really a life worth living?Portrait of Socrates. Marble, Roman artwork (1st century), perhaps a copy of a lost bronze statue made by Lysippos. Current location Department of Greek, Etruscan and Roman Antiquities, Sully wing, ground floor, room 17

Instrumenta propria

Fig. I
HSMoE-AUNet
A microaneurysm is a few dark pixels. A cotton wool spot is a soft pale patch. They do not respond to the same design.
Hierarchical sparse mixture-of-experts attention U-Net on a ConvNeXt-Small encoder, with a different expert set per lesion type. Custom experts: MultiScaleExpert, BrightSpotExpert, EdgeEnhancingExpert, SoftRegionExpert, TextureExpert.
Fig. II
SwinHRUNetPP
Vessel segmentation needs global context and hairline spatial precision at once, and most hybrids trade one away.
A Swin-based high-resolution U-Net++ hybrid, arrived at by rejection rather than by design.
Rejected along the way
- SwinV2-Large
- smp UNet++
- plain SwinUNet
- Lovász loss
- BoundaryLoss
- warm-restart scheduling
- label smoothing
- encoder freezing
Each of these made the result worse.
Fig. III
MultiScaleMANet
Microaneurysms vanish at low resolution and lose context at high resolution.
Three branches — full 512×512 through EfficientNet-B4, four 256×256 patches through a shared EfficientNet-B0, sixteen 128×128 patches through the same encoder — with spatial reassembly and bottleneck fusion. Alignment was proved on paper before implementation.
Fig. IV
RichAttentionUNet
A baseline worth trusting has to be one you can account for line by line.
Written from scratch with no pretrained weights: residual conv blocks with SE channel attention, an ASPP multi-scale bottleneck, attention gates in the decoder, deep supervision heads.
Fig. V
Custom losses
Off-the-shelf losses reward the wrong thing on sparse, bright, or boundary-defined lesions.
DualEdgeBoundaryLoss, OpticDiscSuppressionLoss, SoftBoundaryLoss, probability-weighted RegionCoherenceLoss, an FOV-masked composite loss, clDice.

Axioms

HierarchiRetina, Stage IIIA number nobody can compare with their own is a report about your hardware, not a result.
HierarchiRetina, Stage IThe test set is opened once, after the threshold is locked. There is no second first look.
Renal length, KUB CTMeasure in millimetres, not in slices. The scanner's axes are not the organ's.
Rank radii, unseen cladesA threshold does not survive a change of population. Its quantile does.
Taxonomic placement, refusalRefusal is a prediction. A model that cannot decline will place every specimen somewhere.
HCGT-PG, the estimandWhere measurement is cheap, correlation is free. Only an intervention charges for it.

Elementa

A structured journey through the core mechanisms of modern AI. A course in deep learning, built to be read in order or explored by interest. From foundations and neural networks to bioinformatics and intelligent systems. Each entry is a proposition: a claim, its dependencies, with figures and mathematics.

The whole corpus · The figure library

Marginalia

Reviews of books and models, and notes that did not fit anywhere else.

The whole ledger →

Instrumentarium

Fig. i

Languages

Python (primary). C (coursework level).

Everything below is written in the first of these.

Fig. ii

Deep learning

PyTorch, timm, segmentation_models_pytorch. Mixed-precision training with GradScaler tuning, EMA, test-time augmentation, k-fold CV with resumable per-fold checkpointing, class-imbalance handling (pos_weight, weighted samplers), threshold calibration on validation only, ordinal regression (CORN), mixture-of-experts, cross-attention fusion, attention gates, ASPP, GeM pooling, deep supervision, frozen-backbone transfer, Grad-CAM.

Training discipline more than library familiarity.

Fig. iii

Architectures used hands-on

ConvNeXt V2, Swin UNETR V2, DINOv2/v3, EfficientNet, U-Net, U-Net++, Attention U-Net, SwinHRUNetPP, HSMoE-AUNet, image-text encoders.

I build, train, modify and evaluate these rather than only reading about them.

Fig. iv

Medical imaging

Fundus photography, non-contrast CT, NIfTI and DICOM handling, FOV masking with morphological erosion, CLAHE on the green channel, retinal cropping, patch extraction and stitching, lesion-level connected-component evaluation, de-identification, radiologist report parsing.

Most of this is preprocessing nobody writes papers about.

Fig. v

Evaluation

QWK, ICC, AUC-ROC, AUC-PR, Dice, IoU, NMI, lesion-level and image-level recall, sensitivity/specificity tradeoff analysis, per-dataset breakdowns, held-out test discipline and leakage assertions.

The per-dataset breakdown is the part that matters.

Fig. vi

Causal inference and computational biology

Causal graphs and DAGs, d-separation, interventions, counterfactual reasoning, gene perturbation analysis, gene regulatory network inference, multi-omics integration and biomarker discovery.

Currently extending this direction toward single-cell foundation models, genomic sequence models and perturbation-based causal analysis.

Fig. vii

Data

pandas, NumPy, Albumentations, OpenCV, Matplotlib, Seaborn, scikit-learn, nbformat.

Ordinary tools, used daily.

Fig. viii

Backend

FastAPI, REST APIs, SQLAlchemy, JWT auth, Alembic, AWS S3, CloudFront.

From building CSMMS, in the appendix.

Fig. ix

MODEL FAMILIARITY

Beyond the architectures I have used hands-on, I regularly study modern vision, multimodal, language, medical imaging and computational biology models. This includes SigLIP 2, EVA-CLIP, SAM 3, nnU-Net, ViT, Qwen, Llama, DeepSeek, Kimi, MedGemma and genomic models such as AlphaGenome and Geneformer. I focus on understanding their architecture, training objective, scaling behavior, benchmarks, limitations and practical use rather than treating familiarity with a model as equivalent to hands-on experience. The model notes I keep are part of this ongoing study.

I study, compare and reproduce modern models to understand their architecture, training objectives, benchmarks, limitations and where they fit into my own research.

Fig. x

READING

I love reading books and research papers. My current reading is centered on machine learning, causality and computational biology. Key books include The Book of Why, Causality: Models, Reasoning, and Inference, Causal Inference in Statistics: A Primer, and Elements of Causal Inference: Foundations and Learning Algorithms. The reading is not separate from the research work. It is mainly about understanding what can be inferred from data, where the assumptions enter, and how interventions can provide stronger evidence than observation alone.

I read research papers and technical books to build the theoretical foundation behind the models and methods I use, especially in machine learning, causality and computational biology.

Fig. xi

Other

Git, Jupyter, LaTeX, lex/yacc, CUDA-level debugging.

One machine, a great deal of patience.

Not yet: SQL, propensity score methods, Mendelian randomization.

Chronicle

Fall 2026 —
BRAC University, Dhaka
MSc, Computer Science and Engineering.
Thesis proposal accepted: HCGT-PG. Supervisor: Dr. Badhan Das.
Fall 2021 – Summer 2026
BRAC University, Dhaka
BSc, Computer Science and Engineering.
Thesis: HierarchiRetina, supervised by Rafeed Rahman; not publicly released, and the work is now under review at IEEE JBHI. Coursework in machine learning, deep learning, compilers, theory of computation, DBMS, operating systems, algorithms and software engineering.
2020
Uttara High School and College, Dhaka
Higher Secondary Certificate.
Science.
2018
Ullapara Merchant Pilot High School, Sirajganj
Secondary School Certificate.
Science.

Courseworkmachine learning, deep learning, compilers, Python, theory of computation, DBMS, operating systems, algorithms, software engineering.

Appendix

Systems work: where models stop being isolated learners and become architectures. The chapters above concern what a model learns; these concern how intelligence is assembled across vision, language models, vision-language models, and retrieval through model architecture, multimodal integration, retrieval pipelines, reranking, inference, and evaluation. The discipline is no longer only about learning representations, but about designing systems that can perceive, retrieve, reason, and answer. Each project below opens its own page.

Correspondence

To —

md.asif.uddin@g.bracu.ac.bd github.com/asifuddin01 linkedin.com/in/md-asif-uddin01 orcid.org/0009-0000-1705-4097Dhaka, Bangladesh